Summary: During FY16, we made progress on several subprojects: 1) Herpesvirus Assembly. Over the past 25 years, we have studied many aspects of herpesvirus assembly, focusing on formation of the capsid and its packaging with DNA. In FY16, we completed and published two such studies. The first paper reports a detailed cryo-EM analysis of the procapsid of herpes simplex virus type1 (HSV-1) and the early stages of its maturation. As the procapsid matures, it undergoes major changes in structure and composition. In addition to an internal scaffolding protein, assembly is guided by an external scaffolding protein, the triplex, which coordinates neighboring capsomers. To investigate assembly, we developed a novel isolation procedure for the labile procapsid and collected a large set of cryo-EM data. In addition to procapsids, these preparations contain maturation intermediates, which were distinguished by classifying the images and calculating a reconstruction for each class. Appraisal of the procapsid structure led to a new model for assembly; in it, the protomer (assembly unit) consists of one triplex, surrounded by three major capsid protein (MCP) subunits. The model exploits the departure of the triplexes from 3-fold symmetry to explain the highly skewed MCP hexamers, the triplex orientations at each 3-fold site, and the T=16 architecture. This study also yielded new insights into maturation (see Aksyuk et al. (2015)). In the second paper, the internal proteins of HSV-1 capsids were localized by bubblegram imaging, a novel cryo-EM-based method technique described further below. Analyzed were procapsids (see above) and three mature capsids isolated from the nuclei of infected cells. A-capsids are empty; B-capsids retain a shrunken scaffolding shell; and C-capsids - which develop into infectious virions - are filled with DNA and ostensibly have expelled the scaffolding protein. We found that the scaffolding protein is exceptionally prone to radiation-induced bubbling (this protein forms thick-walled inner shells in the procapsid and the B-capsid). C-capsids generate two classes of bubbles: one occupies positions beneath the vertices of the icosahedral surface shell; the other is distributed throughout its interior. A likely candidate for the subvertex-protein is the viral protease. A subpopulation of C-capsids bubbles particularly profusely and may represent particles in which expulsion of scaffold and DNA packaging are incomplete. Based on these findings, we propose that the axial channels of hexameric capsomers afford the pathway via which the scaffolding protein is expelled. See Wu et al, 2016. 2) Localization of Buried Proteins by Bubblegram Imaging. While many approaches can be used to label proteins that are exposed on the surface of a macromolecular complex, few are applicable to proteins that are buried inside. We have been developing the use of radiation damage in vitrified specimens for that purpose. Sustained exposure to the electron beam elicits the formation of bubbles of hydrogen gas in proteins. The locations of the bubbles can be determined in three-dimensional reconstructions calculated from previously recorded images of the same specimen, then undamaged. Our first success (Wu et al, 2012) was with PhiKZ, a large and complex virus that infects Pseudomonas aeruginosa. This was followed in FY14 with characterization of the internal structure of bacteriophage T7, viewed as a partially defined model system. In the past year we completed a bubblegram analysis of herpes simplex virus capsids (see above) and one of bacteriophage P22 (see below). 3) Localization of the Ejection Proteins of Bacteriophage P22. While it has long been known that capsids serve as delivery vehicles for viral genomes, there is now growing awareness that they also deliver proteins into their host cells. The Salmonella phage P22 has three such proteins (called ejection proteins or E-proteins). Their locations in the virion have remained unknown despite their copious amounts (total: 2.5 mDa) and the generally well characterized P22 structure. Its capsid is a T=7 icosahedral shell with a portal protein dodecamer at one vertex. Extending outwards from that vertex is a short tail, and putatively extending inwards is a 15 nm-long alfa-helical barrel formed by the C-terminal domains of portal protein subunits. We succeeded in localizing the E-proteins by bubblegram imaging. Interestingly, the portal barrel, 15nm-long in a crystal structure, is only about half as long in situ; the remaining, disordered, portion presents binding sites for E-proteins. Thus the E-proteins are loosely clustered in the region radially inwards from the portal crown where they are strategically placed to pass down the shortened barrel and through the portal ring and the tail. These observations (Wu et al., 2016) document a spectacular example of a regulatory order-disorder transition in a supramolecular system. 4) Assembly and Maturation of Bacteriophage Capsids. Our interest in capsid assembly centers on the massive conformational changes that accompany their maturation. These transitions afford unique insights into allosteric regulation. We study maturation in several phage systems. The tailed phages afford an excellent model for herpesvirus capsids, reflecting common evolutionary origins. In FY16, in addition to our investigation of phage P22 (see Wu et al., 2016), we made initial progress on the giant phage S. Giant phages were discovered only recently because their great size limits their diffusability in agarose gels, the traditional phage assay. We find that S capsid has the same T=27 icosahedral architecture as phiKZ (mentioned above) but a very different arrangement of internal proteins, as detected with bubblegrams. This study is ongoing. 5) Papillomavirus maturation. The Papillomaviridae are a family of DNA viruses that inhabit the skin or mucosal tissues of their vertebrate hosts. Unlike other non-enveloped viruses, papillomaviruses are released into the environment through a gradual process called desquamation. During this process, disulfide cross links are formed between neighboring molecules of the major capsid protein, L1. This is thought to stabilize the maturing virion. In FY14, we completed a study in which time-lapse cryo-EM was used to study the maturation of HPV16. Initially, the virion is a loosely connected procapsid that condenses into the mature papillomavirus capsid. In this process, the procapsid shrinks by 5% in diameter, and its pentameric capsomers change in structure (most markedly in their axial region), and the interaction surfaces between adjacent capsomers are consolidated. Our current goal is to characterize next generation virus-like particles intended for use as improved vaccines against cervical cancer. 6) Hepatitis B virus (HBV) is a major cause of acute and chronic liver disease. HBV is a small, enveloped DNA virus whose core gene codes for two variants: core antigen (HBcAg), which assembles into icosahedral capsids; and a precore protein, which is secreted as the nonparticulate e-antigen (HBeAg). HBcAg consists of an assembly domain (residues 1-149) and a C-terminal domain (residues 150-183). HBeAg is directed to the ER via a 29-residue signal peptide. Mature HBeAg consists of the assembly domain plus 10 residues of the signal peptide. Previously we showed that the propeptide induces a radically altered mode of dimerization of HBeAg relative to HBcAg. In FY16 we pursued structural studies in which a scFv was used as a crystallization chaperone, and obtained a co-crystal that diffracts to 1.73 . When compared with our earlier structure solved at 3.2 , the resulting structure shows distinct conformational differences between the paired HBeAg monomers, indicating a flexibility that may have functional implications.